Hi and welcome to our Pensions Age video interview. I’m Laura Blows, editor of Pensions Age and joining me today to talk about data and systems for pension schemes is John Broker, director of ITM Limited.

So John, there has been a lot of focus on data quality in pensions schemes lately, and for administrators of workplace pensions, they’ve got a lot of challenges at the moment, in terms of new regulatory requirements. Now as ITM is at the coalfaces of this, helping pension schemes manage these issues, perhaps you could provide me with an overview of the issues that you are seeing for pension schemes?
Well you are absolutely right, we are very much working at the coalface. I have to say none of these issues, with the exception of auto-enrolment, are new, but it’s only now we’ve found that trustees and employers are addressing these things. I think that’s interesting. A lot of the work we are doing is very much around auto-enrolment, from an admin perspective. And from a data perspective, it’s helping trustees and sponsors get their data in shape to address the challenges we’ve got in this new world of pensions. We’ve got liability reduction, the regulator, lots of different things, different challenges, and that’s where ITM fits in, that’s what we do.

So there seems to be a lot of issues for pension schemes to manage; perhaps we can look at these in a bit more detail. For instance you mentioned auto-enrolment, which is a hot topic at the moment and will continue to be so for a good few years at least. Perhaps you can talk me through what you would advise for those schemes looking at the set-up process for auto-enrolment?
A lot of the big companies have now staged and I’ve heard various commentaries about how they’ve done that and gone through that, and they’ve been incredibly well managed with big budgets, but let’s look at the smaller employers.

Yes, they may be the ones that have a few more problems, or do not have the big budgets to manage auto-enrolment with ease?
I think you’re right. I think the thing is Laura is that people tend to ignore the admin systems element. They first all have to decide what they are going to use as an investment vehicle, difficult choice enough, but what about the administration? What I think people should be focusing on is firstly an evaluation of their current process. How does the payroll work, what’s the HR system, how will all that be brought together to successfully auto-enrol people? And how are you going to communicate with your workforce? So a lot of the work we’re doing is undertaking that evaluation, and then providing middleware through a tool that we use called eAsE. eAsE is set up to bring all of those things together, and help people, through the middleware, to conduct well-ordered and very timely implementation. You need to get a project team together, you need to have that evaluation, and you need to look at your systems infrastructure. And you need to try and understand what it’s going to cost you. So good middleware can do all of these things and communicate with the newly-enrolled employees to tell them that they have been enrolled and what to do about it next.

What kind of timescale would you say pension schemes should be looking at; how much of a big challenge is that?
Sure, these are big challenges as you said. I think my experience is that people leave things too late. And what we would recommend is at least six months before you even think about the actual staging date you need to start your admin prep for that. And that begins with all of the things I’ve just described and it is quite an arduous process. So the message there is to prepare now or prepare to fail.

Strong words there.
Well, that’s the way it is. From experience a good six months is needed and it’s getting the thing on the agenda amongst so many difficult issues for employers to manage in an recessionary climate we have.

So the timescale is looking to be around now for smaller to medium schemes?
Yep, a lot of them will be staging in 2014. I don’t think it’s even on their horizon right now. I know directors are busy people in a difficult climate to operate business in. So now is the time to start this preparation.

So maybe in conjunction with that is another issue you mentioned earlier, which is that of data. Perhaps you could give me an overview on why there has been this focus on data now in particular? As you said it’s an old issue.
It is an old issue; it’s been around years. You and I have seen these things come and go in the years we’ve worked together. The thing that has put this on the agenda was The Pensions Regulator’s new guidance, which came out in June 2010 and was the first time trustees had targets to meet for 2012. In addition we’ve seen the advent of new types of pension solutions with buy-ins, buyouts, pension increase exchanges exercises, lots of things that demand data to be right and right first time. That’s really put this on the map. When we started ITM 10 years ago we were doing much the same work, but much less of it, so all of these things have driven and created a lot more impetus and focus on pension scheme data and that has to be a good thing.

Yes, I’m quite curious though about data quality issues. For instance I assume pension schemes were ticking along fine with the admin side with their data, but now that they’re looking at doing these exercises, they’re finding that their data isn’t actually up to scratch. Why is that?
I think there has been a bit of a ‘head in the sand’ attitude about it. So for a start people have to address it now. The regulator requires you to address that. But whilst we can hold our data to be fit for day-to-day admin, we can muddle along, we can go down into the basement and get the microfiche out, for day-to-day stuff. When it comes to a liability reduction project, our data has to be on that system and it absolutely has to be correct. Let me give you an example. A lot of organisations will come to us and say ‘is our data ready’ to complete a buyout for example or a pension increase exchange exercise. The key things you need to be looking at are your GMPs correct and are your pension splits correct between the pre and post 1997 accrued amounts. Regrettably the state of the nation is they are not always as good as they could be. So that creates a big exercise to clean that data up and that has never been addressed before. So there are shocks for trustees and sponsors in terms of time, effort and cost, all of the things you need to do to get ready to transact.

Perhaps you could talk a bit more about the issues you found with these companies going about a data cleansing exercise. What practical issues have there been?
First of all there is a reluctance to address it because let’s face it, it’s not the most exciting topic out there amongst the array of things we can do in terms of pension products.

Is there a fear of cost issues as well?
Definitely. People need to understand the benefits weighted against the actual costs of doing such an exercise. Let me give you some examples. An organisation transacting a buyout recently; we found that up to 30 per cent of their records were incorrect. By going through the data analysis and data cleanse preparation process we saved them a considerable amount of money. I’m talking millions of pounds in reduced buyout premium. But it’s not just about cost; it’s about de-risking your data. So an understanding of what you’ve got there. And there’s an assumption, which is arguably incorrect, that ‘our data will be alright, after all, we’ve been getting by for years, why now is it a problem’, and so it’s a shock. And the cost of cleaning can be quite a lot but it’s absolutely nothing compared to the saving you’ll get back by addressing it.

So what other benefits could a pension scheme achieve by undertaking this data cleansing?
For final salary schemes it is generally acknowledged that legacy data is not as good as it could be. So the benefits are to the trustees. After all, they have a duty to pay the right benefit to the right member at the right time. So if they have been through this exercise they have some certainty that they can discharge their fiduciary obligations much more accurately. There is a lot more certainty, and perhaps comfort, around that. From an employer’s perspective, it will enable them to transact a liability reduction project. So for DB there is a clear case, and clear benefit. For DC, often data is better, because these schemes are newer and we’ve not had quite as much time to get it wrong. However the same principles apply. It’s about how good is my data, can I discharge my fiduciary obligations correctly and what do I need to do to make sure and have that comfort.

Moving on from data issues itself, there are a lot of other challenges that administrators have to look at. For instance there is the issue of GMPs and the announcement we are waiting for about how they can be converted into a scheme benefit. What do you think about that, is it realistic and what should trustees and employers be looking at to handle this?
I think the DWP have announced that in 2014 what we will have to do is equalise GMPs. and then convert them into an ordinary scheme benefit. A worthy aim perhaps but it’s going to be a headache. A think what we are going to see is a rush on HMRC time, because before you can undertake those equalisation calculations, the GMPs have to be present and correct to begin with. So now is the time to do your GMP reconciliation and getting them right in preparation for the equalisation if it goes ahead on time as the DWP has said it will do. I think the actual methodology for converting into ordinary benefits is very vague at the moment so we are expecting more direction. So I think sponsors and trustees are going to welcome that.

So in the meantime, is it ‘wait and see’, or can you start preparing?
You need to prepare now, and you need to build that into a good data project, where a GMP focus is high on the agenda.

So how can they build this into their data process? How can you ensure that your data is up to date for all of these issues?
This starts at a slightly higher level. I think what people need to do is take a moment out and think ‘what are we doing with the scheme? Where are we going with it? Are we on a flightpath to closure? Liability reduction all the way through?’ There are a number of considerations there. Data, admin, communication, costs, funding. So the starting point has to be to think about all the things that we want to do with our scheme and fit our data project objectives to meet those as they occur over the timeline that is envisaged over the short, medium and longer term.

So for a lot of pension schemes that’s probably quite a number of big projects and challenges to deal with. Is there software and support available to help them with this? Has the software caught up with the issues that pension schemes are facing?
That is a bit of a bugbear of mine.

We’re not known for being modern with technology, which is an issue I think.
It is an issue, and if we look at other financial services industries, like online banking, they have online everything. Banks are not exactly flavour of the month but they do provide good, online availability. There is a reluctance to invest in good systems. Finance directors will naturally think ‘why should I spend money on good technology? Why should I spend pensions money on a new admin package or auto-enrolment system?’ My view about that is that we just have to do this. We have to move with the times. We’re all involved in managing pension schemes, we all have an obligation to each other and to members to make sure that our automation comes up to speed. So they do not have to cost the earth, there are really good ways of implementing cost effective technology for auto-enrolment, pensions admin, or data management. So my view on that is that we must move with the times and create budget. Invest now. Don’t fall further behind.

So it seems that pension schemes may have many projects to deal with, but the software and support is available to help them with this and they are able to go to companies to help with this.
Definitely. There are a number of great products out there, including our own. Again it is very much getting that focus on it. It is a difficult climate, with FDs very much focusing on funding, and matters such as actuarial investment. The admin side of it, technology side, data, does tend to get relegated to below that radar. It’s often not the most exciting project and spend of money. But if you look for example at the number of complaints TPAS receive relative to pension admin, it’s very interesting that it grows year on year. Now is it because people are more aware of these issues and so complain more regularly or is it that we just haven’t got this right still. A good measure arguably of good admin is to see a number of reductions in the complaints that TPAS is receiving, but we’re not seeing that. A lot of these things are about basic data system failure and errors.

So data is rising up the agenda in terms of importance for those managing pension schemes, but yet it still needs to be raised higher still?
It’s going up. We’ve had this focus from the regulator on common data, which is fairly basic stuff, and that’s very good, we welcome that, as most people in the admin community would do. It’s created good focus and has gone on the trustee agenda. The regulator is now monitoring compliance with that issue, so it will be interesting to see what comes out of that and how tough a stance they take on it in terms of policing it, and with trustees that have been errant and not complied, just how hard a stance they will take. So it’s very much something that is going to come to the fore in the press very soon. But more importantly the regulator has decided that it is going to focus on conditional data. To me, that is a very good initiative. We are actually going to have a focus on the data items that make the benefits tick, make them work and make them fit for purpose. Now that’s going to be a very expensive and very time consuming exercise. But it’s absolutely the right focus, so the regulator has got this spot on.

So improving the quality of the data will end up improving the quality of the scheme overall for members?
Whether it be for auto-enrolment or whether it be for administering a pension scheme, clearly having the right data has a lot of benefits. Let’s look at some of them right now. First of all, good data is good management information. Being trustees is like running a business; you need good MI, your liabilities reflected in your data. What about funding? If your data is right your liabilities are likely to be in better shape than they were before you put your data right. So one could argue that your funding will be more accurate. And for the members and day-to-day admin, for the administrators it’s made a lot easier. It is a difficult and thankless task administrating pension schemes. So ultimately the key thing is that members will get better service. It’s not just about the speed of service, it’s about accuracy. I would very much like to see pension admin standards rising and to see those TPAS complaints falling.

So it looks like things will be changing for the better hopefully.
I think so. I think we’ve got a way to go and it is difficult to do all these things. Sponsors and pension fund trustees have lots of competing priorities and limited budgets but for once we’ve got the right focus on the right issues from an admin perspective.